RSS
Christopher F. Marki received his B.S.E.E. from Duke University in 2002 and his M.S.E.E. and Ph.D. from University of California, San Diego in 2004 and 2007, respectively. While in graduate school, Christopher studied high speed fiber optics and consulted for San Diego start-up Ziva Corporation. Following graduate school, Christopher decided to forego a life in Photonics and opted, instead, to work with his father at Marki Microwave and learn the “family business” of microwave mixers. While at Marki Microwave, Christopher has served as Director of Research and has been responsible for the design and commercialization of many of Marki’s fastest growing product lines including filters, couplers and power dividers. Dr. Marki has authored and co-authored numerous journal and conference publications and frequently serves as an IEEE reviewer for Photonics Technology Letters and Journal of Lightwave Technology. MarkiMicrowave.com
To comment or ask Christopher a question, use the comment link at the bottom of the entry.
July 2, 2010
I love soccer (i.e. football for my international readers). I grew up playing the sport and consider it one the most character-defining experiences of my life. For me, the World Cup is the greatest sporting competition around. Now that I am in the technology area, I am dumbfounded at FIFA’s insistence on ignoring modern technology to improve the officiating. Just like the rest of the world population, I find FIFA’s stoicism ridiculous and alarming when the fate of entire nations (and millions of dollars) rests in the hands of one or two terrible refereeing mistakes (hello
With Sepp Blatter’s recent announcement that FIFA will “re-evaluate” the use of goal line technology, I have started to imagine how technology could transform the officiating of the world’s game. I am no futurist, but this thought experiment provokes some interesting questions as to what is currently feasible with modern technology, and what still requires some R&D. The following analysis follows similar guidelines to the way I evaluate new product development and directions at Marki Microwave.
As one might expect, the low-difficulty, low cost solutions are the areas that FIFA should adopt first. The blue-sky areas (high difficulty, high cost) might never be considered for any sport, but are interesting talking points that would make any Venture Capitalist salivate.
Sideline/Goal-line Technology
Difficulty: High School Science Fair
Cost: Minimal
Technological Requirements: This is a no-brainer. All you need is an array of cameras, some image recognition software that already exists (such as the Hawk-Eye system in Tennis), maybe a few RFID tags embedded in the ball, and a big red light that flashes when the ball cross the line (like in my other favorite sport, Hockey). The fact that these technologies already exist in other sports but have been stubbornly ignored by FIFA has nothing to do with science, I’ll let the bureaucrats fight this one out.
Automatic Offsides
Difficulty: Undergraduate Research Project
Cost: Minimal
Technological Requirements: I would love to see this employed. It seems that as long as there are a few cameras at high enough angles in the stadium, real time software can easily determine the position of the forward most attacking player at the instant the ball is played. Amazingly, TV broadcasters are already using a variant of this technology (albeit after-the-fact) to determine whether the call was correct or not. If we really wanted to get elaborate, we could use an antenna array and RFID chips to triangulate the exact position of the ball and players. We could then place an accelerometer in the ball to determine the exact moment the ball is played. All of these technologies would be synced to some kind of central processor that could easily determine if the player was offside or not at the exact moment of impact. This can be done with off-the-shelf products (I’m guessing for less than the cost of a plane flight to
Diving Detection
Difficulty: PHD Dissertation
Cost: Moderate
Technological Requirements: For me, the most unappealing aspects of soccer are diving and injury faking. Compared to other sports like Hockey, where diving is rare and playing through injury is commonplace, soccer is full of primadonna stars that don’t like to get their shirts dirty (did someone say Cristiano Ronaldo?). Nothing makes me happier than when a player is carded for diving in the penalty box. It appeals to my sense of justice.
How can technology be used to clean up the game? This is a difficult question because computers are not good at subjective decision making. In many cases, diving does involve some amount of contact with a defender. Therefore, a first requirement is that we would need a good array of cameras to follow the play from multiple angles to determine how much contact is made. We could then use some kind of physics modeling engine (like those used in video games) to predict reasonable outcomes from the contact. In the cases where totally unreasonable outcomes occur (like when a player clearly dives without being touched), the computer could notify the referee that an unpredicted or unreasonable result occurred based on the forces involved in the tackle.
The hardest part, however, is that diving is often contextual. Players tend to dive in certain parts of the field, and some do it more than others. We then require that the software learn to incorporate some kind of learning algorithm such that it can develop a “soccer sense”. For example, we know that Cristiano Ronaldo is likely to dive when he dribbles straight into 3 defenders. Programming soccer sense would be one of the most difficult problems for computer scientists. Conveniently, governments, corporations and universities are interested in this kind of computer learning and decision making because we increasingly rely on software to predict our moods and behavior given a set of initial conditions. The ability to “train” computers to understand and interpret human behavior is incredibly important for many fields including Marketing, Security and Investing, so I’d imagine one can extend it to Sport.
Human-less Officiating
Difficulty:
Cost: Millions in Venture Capital
Technological Requirements: Given enough resources and the right collection of experts in computer science, robotics and networking, this might be possible in a few years. It all comes down to the software’s soccer sense algorithm. A human referee must account for many conflicting variables during a game: score, time left, emotional state of the players. It is the nuances of the game that make it so entertaining. Sometimes the correct decision is to allow the game to be more physical, sometimes not. Sometimes a card is warranted, sometimes not. The decision making is contextual, I suspect this would be difficult to teach to a computer without significant R&D. But, it is not impossible. It would just take lots of patience and even more calibrating and tweaking. For all we know, someone at Google or IBM is already doing this…
It looks like the goal line technology is quickly on its way to adoption based on public outcry. However, I don’t see FIFA making any other changes any time soon. As a fan and technologist, I just want to see an equitably called game free from referee tampering. The